INTRODUCTION
Evaluation on the production of different Lotus spp. grown at the level of an introduction garden in our Institute was previously presented (Remis et al., 1995; Locatelli et al., 1997). The introduction garden is located in the so called Salado River Basin, in Chascomús County, close to the Chascomús Pond (latitude: 35º 35' South; longitude: 58º 00' West). The region mainly dedicates to beef and dairy cattle production by using a grazing system essentially based on natural grasslands, with a minor portion of cultivated pastures. Soils in the area characterize by a high salinity and alkalinity, being poorly drained and periodically exposed to waterlogged conditions. All these conditions significantly decrease persistence and yield of common legume species in the region (Mazzanti et al., 1986), turning Lotus spp. as an important alternative for the use as a forage. Aims of our studies are basically oriented to the evaluation of adaptability, of several Lotus spp., to the soil and environmental conditions of the region and looking for the selection of species of the higher forage potentiality. Our previous works (Remis et al., 1995; Locatelli et al., 1997) reported on different characteristics related with the performance of Lotus spp. grown in the region and on variations of the nutritional quality of the forage throughout the year.

Multiplication of seeds from species showing better behavior is a key step for the introduction of a new forage. In this way, it is possible gradually increase the stock of seeds in order to develop assays in higher field extensions. In the present work we further characterize Lotus spp. exhibiting good growing performance in the region, describing phenological and nutritional properties of the plants at reproductive stage. Morphological and sanitary characteristics of the produced seeds are also reported.

MATERIALS AND METHODS.

The different Lotus spp. evaluated produced flowers and fruits during the period spring-summer of 1996-1997. To characterize nutritional quality of the forage at the reproductive stage (middle of December, 1996), samples of leaves, stems, flowers and immature pods were harvested. Samples were dried in an oven at 64ºC until constant weight in order to determine dry matter (DM) content. After crumbling of samples to 1mm, crude protein (CP) and neutral detergent fiber (NDF) were determined by the method of Kjeldahl (N x 6.25) and after Goering and Van Soest (1970), respectively.

Mature pods were harvested manually during the first half of January, 1997, when about 80% of them changed from light green to brown color. Samples were placed on nylon at room temperature in order to eliminate excess of humidity. For each species of Lotus it was analyzed the most frequent number of pods per umbel and on 20 randomly selected pods it was settled the mode of seeds per vaine. The weight of 1,000 seeds was also determined.

Attack of Bruchophagus platypterus to seeds and the evaluation of the infection was carried out after Mujica (1987).

RESULTS AND DISCUSSION
Flowering of the species L. filicaulis, L. corniculatus var. hirsutus, L. corniculatus var. corniculatus, L. glaber (syn. tenuis) (naturalized), L. uliginosus (syn. pedunculatus), L. corniculatus x uliginosus and L. corniculatus var. arvensis started in the first half of November and reached a maximum one month later. These plants fructified in the first half of January. For L. rectum, flowering was more concentrated and finished before (from 11/20 to 12/ 20) with the formation and ripening of pods produced by the middle of January. Except for L. rectum, all the other species performed a second flowering period extended to the end of February, but pods in this case did not reach complete ripening even by the end of March, probably because of changes in photoperiod and temperature by this time. Results suggest that production of seeds of the different Lotus spp. reaches optimum by the month of January.

Table 1 shows evaluation of nutritive quality of different Lotus spp. as a whole forage (flowers, leaves, stems and fruits) and at the reproductive stage. Naturalized L. glaber was included as a control, since this species spontaneously grows close to the introduction garden. In agreement with previous reports (Echeverría et al., 1986) Table 1 shows that L. glaber exhibited a higher quality (higher protein and lower fiber content) than the other species evaluated. As shown in Table 1, L. uliginosus produced a forage with high levels of protein but also with high values of fiber; whereas L. corniculatus var. hirsutus and L. filicaulis exhibited contents of about 13% of protein and high levels of fiber.

Table 1. Nutritive parameters in whole plants of Lotus spp. at reproductive stage.

Species	g DM/kg GMa	g CP/kg DM	g NDF/kg DM
L. glaber (naturalized)	257	184	417
L. corniculatus var. hirsutus	207	137	nd.
L. uliginosus	164	173	513
L. filicaulis	215	126	575

^aGM. Green matter.

Characteristics of pods and seeds of different Lotus spp. are included in Table 2. As shown, L. rectum distinctively exhibited the higher number of pods/umbel, with a low number of seeds/pod, thus showing particular morphologic characteristics for this species. It has been reported (Mlynarczyk and Olesinski, 1993) that different varieties of L. corniculatus show umbels containing 5 or 6 pods, data in agreement with our results in Table 2, except for L. corniculatus var. corniculatus (variety from Turkey) with only 2 pods/umbel. Although it is known that the number of seeds/pod usually is a variable parameter (i.e. for L. corniculatus is between 4 and 15, see Mlynarczyk and Olesinski, 1993), our data show that the different Lotus spp. presented 10-15 seeds/pod, except for L. rectum with the lower number and L. corniculatus var. hirsutus with the higher number of seeds/pod (Table 2). It is worth noting that this later result is in good agreement with the distinctive high reseed capacity exhibited by L. corniculatus var. hirsutus.

Seed size is associated with seedling vigor, being this a critical topic for Lotus spp. (Beuselinck and McGraw, 1983). Our data on the weight of 1,000 seeds showed in Table 2, agree with previous reports on values for this parameter between 1.0-1.5 g for L. corniculatus (Mlynarczyk and Olesinski, 1993). As a whole, results in Table 2 show L. corniculatus var. hirsutus as a variety exhibiting a good performance in terms of the number and quality of seeds produced. This reinforce the potentiality of L. corniculatus var. hirsutus, which also exhibits high productivity and nutritional values, as previously reported (Remis et al., 1995).

Table 2. Morphological attributes of pods and seeds from Lotus spp.

Table 3 shows results on the evaluation of damage of seeds of Lotus spp. caused by infection by B. platypterus in three consecutive years. As shown, in 1995 infection by trefoil seed chalcid was higher in all the different Lotus spp. (except for L. rectum) (Table 3), and damage increased during the summer period (data not shown). These results are in the range of values obtained by Peterson et al. (1991) in Wisconsin (damage of 1.7-39.2 %), by Nang'ayo (cited by Peterson et al., 1992) in Ontario (damage of 1-23 %) and by Mujica (1987) for L. glaber in the Salado River Basin (damage of 0.2-24.1 %). It is worth to point out that in our region, the summer of 1995 was particularly dry and hot, which could explain the high occurrence of B. platypterus. In 1996, damage caused by the bug decreased (Table 3) and reached values similar to those found in the introduction garden in January 1994 (Remis et al., 1995). The following year, seeds of the different Lotus spp. were early harvested, and no infection by B. platypterus was observed in January 1997 (Table 3).

Results showed in Table 3 agree with the high variability between years of the damage caused by trefoil seed chalcid, as reported by Peterson et al. (1991). This is related with weather conditions (mainly temperature and humidity) and with the amount of non-harvested seeds remaining in the field. From our results, it is important to note the high resistance exhibited by L. rectum to infection by B. platypterus along the 3 years analyzed and especially in 1995, when attack of the bug to other Lotus spp. reached a maximum.

In summary, the present work reports on the behaviour of different Lotus spp. at the reproductive stage, in our introduction garden. Phenological and nutritional properties of plants and morphological and sanitary characteristics of the seeds agree with our former reports (Remis et al., 1995; Locatelli et al., 1997) and reinforce that previously suggested about the potential importance of L. corniculatus var. hirsutus for the region. Moreover, results point out the good performance exhibited by L. rectum in terms of high resistance to the infection by B. platypterus, as well as the morphologic characteristics of the seeds produced by this species.

Table 3. Degree of damage (% damge = D) produced by B. platypterus on Lotus spp. seeds.

Species	1995a						1996						1997
	Ib	II	III	D				I	II	III	D		I	II	III	D
L. corniculatus var. hirsutus	-	-	-	-				6	0	0	1.7		0	0	0	0
L. corniculatus var. corniculatus	75	0	6	25.6				5	0	1	1.9		0	0	0	0
L. corniculatus var. arvensis	25	20	0	13.6				-	-	-	-		0	0	1	0.3
L. corniculatus var. alpinus	-	-	-	-									-	-	-	-
L. corniculatus x uliginosus	47	2	2	15.1				-	-	-	-		0	0	0	0
L. uliginosus	-	-	-	-				8	1	1	3.2		0	0	0	0
L. rectum	0	0	0	0				0	0	0	0		0	0	0	0

^aData correspond to seeds harvested in January.

^bNumbers I, II, and III indicate empty seeds with perforated envelope, seeds bug larvae, and seeds containing adult bugs, respectively (see Mujica, 1987).

ACKNOWLEDGMENTS.

The authors wish to thank to José L. Burgos, Ricardo L. César and Mónica L. Sosa for helpful technical assistance. This work was supported, in part, by grants from the Comisión de Investigaciones Científicas (CIC, Buenos Aires). MAA and AAI are Members of the Investigator Career from CONICET (Argentina).

REFERENCES.

Beuselinck, P.R., and McGraw, R.L. 1983. Seedling vigor of three Lotus species. Crop Sci. 23: 390-391.

Echeverría, D., Vernli, C., and Cosio, F. 1986. Características nutricionales de una pradera naturalizada de lotara de hoja angosta (Lotus tenuis Wald et Kit). II. Variación en la calidad de las plantas en el tiempo. Agric. Técnica 46: 245-252.

Goering, H.K., and Van Soest, P.J. 1970. Forage fiber analysis (apparatus, reagents, procedures and some applications). Agric. Handbook 379. USDA.

Locatelli, M.L., Remis, J.L., and Iglesias, A.A. 1997. Evaluation of forage quality attributes for Lotus spp. grown in the Salado River Basin. Lotus Newslett. 28: 17-21.

Mazzanti, A., Darwich, N.A., Cheppi, C., and Sarlangue, H. 1986. Persistencia de pasturas cultivadas en zonas ganaderas de la Pcia. de Buenos Aires. Rev. Arg. Prod. Anim. 6 (Supp. 1): 65.

Mlynarczyk, K., and Olesinski, L. 1993. Evaluation of ecotypes of birdsfoot trefoil (Lotus corniculatus L.) and marsh birdsfoot trefoil (Lotus uliginosus Schk.). Proc. XVII Intern. Grassland Congress (Nice), 235-236.

Mujica, M. 1987. Presencia de Buchophagus platypterus en la República Argentina. Infestación de semillas de Lotus tenuis. Rev. Fac. Agronomía UNLP 63: 82-90.

Peterson, S.S., Wedberg, J.L., and Hogg, D.B. 1992. Effect of cutting and burning birdsfoot trefoil grown for seed on the trefoil seed chalcid (Hymenoptera: Eurytomidae). J. Kansas Entm. Soc. 65: 292-298.

Peterson, S.S., Wedberg, J.L., and Hogg, D.B. 1991. Postdiapause development and phenology of the trefoil seed chalcid (Hymenoptera: Eurytomidae) and its parasitoids. Entm. Soc. America 20: 1606-1611.

Remis, J.L., Ruíz, O.A., Ugalde, R.A., and Iglesias, A.A. 1995. Evaluation of Lotus spp. growth in the Salado River Basin. Lotus Newslett. 26: 17-20.

Back to the Table of contents